Point of View

These columns of ICJ offer an opportunity to the engineering

fraternity to express their views on the current practices in
design, construction and management being followed in the
industry.
To share your opinion with our readers, you may send in your
inputs in about 1500 words via e-mail to editor@icjonline.com

Concrete mix revisited

P.M. Deshpande

Concrete, as on date, remains the most widely used

man made construction material due to its versatility.
The advent of ready mix concrete, have changed some
aspects of concrete. Now, the consumer can specify
the concrete of his needs without bothering about the
ingredients. In the early days, concrete quality was
being controlled by quality of ingredients and by the
concreting process. This generally resulted in uniform
quality of concrete. The development of finished product
quality criteria such as strength, permeability, etc lead
to the neglect of the rigorous process control. This
has resulted in greater variation in properties. These
variations are being taken care of statistically. Thus,
greater the standard deviation, greater is the target
strength. The codes recommend a margin of 10 to 12
with a confidence level of 95%. The six sigma principle
gives lowest strength of 19 and highest one as 61 for M40
concrete. Thus, the variation is 42Mpa. Naturally, this
results in very high consumption of cement. The high
consumption of cement not only affects the economy
but also the environment.
If one analyses concrete strength results of a project
round the year, then one can notice that there are seasonal
variations, variations due to typical location of structure,
and there is variation in day and night concreting. The
concrete quality is also affected by change of source

of ingredients. Apart from variation in individual

characteristics of ingredients and their interaction, the
variation in storage of ingredients, variation in mixing,
variation in composition, can cause wide variation in
concrete quality. Present codal provisions require that
with every such change a new mix design be prepared.
Nowadays, the volume of concreting is such that no
one can stop the work. At the same time, no one can
ensure uniformity of material quality. Such variation
is now taken care by specifying higher standard
deviation. It generally so happens that at the time of
making trial mixes, good quality material is used but the
material at site loses its quality due to storage, etc. Such
material results in bad quality of concrete. Apart from
ingredients, a number of material mixing, transportation
and compaction processes are involved in concrete. It
is seen that optimum quality of concrete is achieved by
pairing best combination of ingredients and processes.
Thus, if there is more silt in the mix, then the ill effects
of silt on quality of concrete can be avoided by extended
mixing of concrete. So, design of concrete mixes has
to be an integrated process involving correct choice of
ingredients and corresponding processes.
The foremost process involved in production of concrete
is mixing of the ingredients. The intimate mixing of
ingredients requires making individual ingredients to

MAY 2008 The Indian Concrete Journal

29

Point of View
be free of their own bond. Thus, it is seen that cement
stored in vertical silos often form lumps, particularly in
humid and cold environment. Sometimes, they do not
dissolve in water even in the presence of deflocculating
admixtures. So, the water cannot reach to inner particles
of cement in a lump, with the result that those particles
cannot be hydrated and part of concrete remains weak
and overall strength is reduced considerably. In trial
mixes this eventuality is not accounted for. Similar is the
case with silt or clay. It is seen that clay or silt in finely
divided state helps making concrete better. If it is in
the form of lumps, a lot of water remains locked in the
lumps. The breaking of such lumps during compaction,
more often releases the water which apparently is seen
as bleeding of concrete.
Similarly, the process of transporting, requires different
workability for different modes of transport. Pumped
concrete requires a cohesive mix with a slump of 8 to
10 cms as a less cohesive mix with high slump (15 to
20cms) is difficult to pump.
At our sites, the aggregates are typically stored in the
open. The aggregates are exposed to stone dust during
production and also to the dust created by construction
activities. Once accumulated on the aggregate surface,
the dust particles are difficult to remove. These particles
cause blinding of pores on the surface of aggregate and
prevent water absorption of aggregate. If extra water is
added to the concrete to cater for water absorption of
aggregates, then such blinding gives rise to additional
free water causing additional slump. These dust particles
then catch water in the concrete mix and feed the pores
in aggregate which causes reduction in the free water
in concrete resulting in loss of slump. If the aggregate,
before mixing, is dry and if the percentage absorption of
aggregate is high, then there is considerable loss of free
water resulting in drastic reduction in workability. In
such cases, cement particles at the surface of aggregate
become thirsty for water and cannot be hydrated. If the
dust contains chemically active clay particles, then the
problem is further aggravated.
It may be noted that during mixing in the mixer, the
surface dust is picked up by watery cement particles
and these dust particles form an emulsion along with
the cement particles and coat the aggregate surface
uniformly resulting in cohesive green concrete and
higher strength hardened concrete.
An experiment was conducted to see the effect of dust
particles and amount of free water. All in one aggregate
without - 600 micron, and water in the ratio of 12:1 (by

30

The Indian Concrete Journal MAY 2008

weight) was taken. The water was added to the aggregate

and stirred. It was seen that there was no free water that
can come out. Then water was increased to 12 :1.5 ratios
and again there was no free water. Subsequently, water
was increased to 6:1 ratio and at this water content, free
water was noticed. The free water was removed and the
quantity of held water was equivalent to the ratio 20:1.
Thus quantity of held water was considerably reduced
when dust from aggregate surface was washed out.
The difference in water content in the 20:1 ratio and 8:1
ratio was held by dust and when the quantity of water
increased, it washed away the dust and the held water
became free. In terms of W:C ratio, clear aggregate holds
water equivalent to W:C ratio of 0.21, whereas the water
held by dust coated aggregate was 0.525. This holding
capacity kills slump and also reduces strength. This is
the reason that sometimes W:C ratio of 0.5 gives higher
strength than that with 0.35. If stone dust is replaced
with clay, then water held would be 0.6 in terms of W:
C ratio.
The effect can be statistically evaluated by designing
an experiment. The experiment involves casting and
testing of trial mixes. The mixes were designed with
two levels of W:C ratios as shown below. Both clean
aggregates and aggregates coated with dust or clay were
used. The mixes so cast were then tested for slump and
compressive strength.
Table 1. Design of experiment for evolving relationship
between dust coating, w:c ratio, and slump and
compressive strength of concrete
Dust on surface

W:C ratio

0.35

0.5

Clean, 0%
With dust coat, 100%

(1)

ab

Table 2. Laboratory test results of Table-1 experiment

Treatment

Slump, mm

Compressive strength,
Mpa

(1)

40

55

110

32

10

33

ab

90

43

The results give the following relationship:

Y1 (slump) = bo + b1 x1 + b2 x2 + b3 x1 x2
= 62.5 + 37.5 x1 12.5 x2 + 2.5 x1 x2

Point of View
where, x1 = (W -0.425)/ 0.075, x2 = (S -50) / 50

The analysis yields the following relationship:

Y2 (compressive strength) = 40.75 + 5.75 x1 2.75 x2 +

8.25 x1 x2

Y = 8.95 - 7.885 x1 - 4.4 x2 + 2.23 x3 + 4.27 x1 x2 - 2.1 x1 x3

1.63 x2 x3 + 1.55 x1 x2 x3

When water and cement is mixed together, the watercement paste is formed. The consistency of the paste
varies as per the proportion of cement and water. It
forms a plastic mass if water to cement ratio is below
0.35. As the water content is increased, it results in
viscous liquid. A further increase results in a flowing
mix. In the plastic mass, water is held by cement particles
whereas in case of flowing mix, the cement is emulsified
and forms emulsion in water. The addition of admixtures
leads to further lowering of paste consistency. The paste
determines the slump of concrete. The lower consistency
results in higher slump but of a non cohesive type.
The medium consistency results in higher slump but
a cohesive mix. The addition of particles of size less
than 300 micron further increases the consistency. The
properties of concrete is dependent on how good the
coating of aggregate surface is achieved by the cement
water paste. The thickness of coating and its uniformity
account for concrete workability and concrete strength
apart from W:C ratio, aggregate and, type and grade
of cement.
The variation in consistency with variation in admixture,
water and silt content can be numerically established by
the following statistical experiment:
Table 3. Design of experiment for evolving relationship
between silt content, w:c ratio, admixture dosage and
consistency of of cement - water paste as measured
with Marsh cone
Silt content in terms of
percentage of cement (S)

W:C (W)

0
10%

Admixture dosage in
terms of percentage of
cement (A)
1%

2%

0.35

(1)

0.6

ab

0.35

ac

0.6

bc

abc

The trial mix results were:

Table 4. Laboratory test results of Table-3 experiment
(1)

ab

ac

bc

abc

18.00

0.97

7.00

33.00

0.85

1.40

9.30

1.00

where, Y = Time taken in minutes to empty the Marsh

cone (measure of consistency)
x1 = (8W - 3.8), x2 = 2A -3 , x3 = 0.2 S -1
The workability or slump of concrete is related to the
consistency of cement - water paste, ambient temperature
and aggregate water ratio. To find out the numerical
relationship, the following experiment was designed.
Table 5. Laboratory test results of the experiment
designed for evolving relationship between dust
coating, aggregate - water ratio, ambient temperature
and consistency of water - cement paste with slump of
concrete
Aggregatewater ratio
(Ag/W)

Ambient
temperature (T),
o
C

2 minutes

8 minutes

10

25

15

40

12

25

10

40

15

Paste consistency (PC)

The analysis yields the following relationship:

Y (slump) = 4.31 0.625 x1 - 1.37 x2 x3 +.0.2 x1 x2 - 0.5
x1 x3 + 3.75 x2 x3
where, Y = slump in cms, x1 = (T - 32.5) / 7.5, x2 = (PC
-5) / 3, x3 =(AW 12.5)/2.5.
The concrete specifications generally stipulate maximum
and minimum permissible values for the various
ingredients. This includes the values of minimum and
maximum W:C ratio, cement content, admixture content,
cement strength, etc. The above equations enable one to
find combined limit from individual ingredient limits.
Thus, limits for cement water paste consistency can be
determined from limits of W:C ratio, admixture, and silt
content of mix. Similarly limits regarding slump, paste
consistency and temperature will enable to find out
the limits for Ag:W. These derived limits will enable to
design experiment to find concrete strength in relation
to various factors like slump, cement content of mix,
cement strength and water absorption of aggregate.
This relationship can be evolved by trial mix with a

MAY 2008 The Indian Concrete Journal

31

Point of View
desired initial slump. Then extreme cement content
limit, available cement strength and aggregates can be
worked out. The experiment may be designed in the
following manner:
Table 6. Design of experiment for evolving relationship
of slump retention and development of strength of
concrete with aggregate absorption, cement content of
mix, and cement strength
Aggregate
absorption, %

Cement
content, kg/m3

2
5

Cement strength, N/mm2

53

67

350

(1)

425

ab

350

bc

425

ac

abc

batching, mixing, transportation, placing, compaction

and curing. The final mix design is done as per
stipulations and tolerances provided in the respective
specifications. The specifications generally provide for
minimum and maximum W:C ratio, cement content,
ambient temperature, etc. The acceptance is defined
by minimum workability, strength, permeability, etc.
The final trial mix is done at the plant using acceptable
tolerances, acceptable concreting process and acceptable
ingredients. For this purpose, statistical design of
experiment considering critical specifications, their
natural variation and acceptability of final concrete is
done. The relationship derived on that basis can be
used to monitor the concrete quality. The relationship
can also be used to simulate the effect of the variations
on the concrete.

Table 7. Table format for recording experimental

results of Table-6 experiment
Treatment

(1)

ab

ac

bc

abc

Strength
(Immediate)
Slump
(Immediate)
Strength
(Cast after 2 hrs
of mixing)
Slump
(After 2 hrs)

The analysis of the above data would then yield the

relationship. Such experiments can be conducted to
see the influence of variations in ingredient levels or
any of the processes on any of the characteristics of
concrete. These relationships enable to achieve optimum
concreting. Thus, if we consider ambient temperature,
it enables to compare the effect of temperature on
strength and slump and thereby the cost of processing.
Similarly, mixing, transportation and compaction can
be optimised by considering cost of ingredients and cost
of process. Thus, the effect of silt in aggregate on the
strength of concrete can be taken care of by removing it
by washing and also by extending the mixing time and
using additional water. Alternative processes can be
chosen by optimising the cost.
The exercise so far has helped in making a choice of
the ingredients, determining the need to process the
ingredients and identifying the concreting process to
be adopted. The concreting process include method of
storage of materials, method of processing ingredients,

32

The Indian Concrete Journal MAY 2008

various
control.

Mr. P.M. Deshpande is B.E. (civil) and has

done P.G. Diploma in Statistical quality control
and Operations Research from Indian Statistical
Institute, Kolkata. He is former Chief Engineer
and General Manager (Tech), CIDCO Ltd. His
areas of interest are design of interactive and
integrated QA systems and investigations of
cause-effect relationships for effective quality

Be an ICJ Author
We at ICJ offer an opportunity to our readers to contribute
articles and be a part of a big family of ICJ authors.
In particular, we will appreciate receiving contributions on
the following:
Articles bearing on innovative design and
construction
Articles dealing with challenging construction
problems and how they were solved.
Just a Point of view covering your opinion on any
facet of concrete, construction and civil engineering
All contributions will be reviewed by expert Editorial
Committee. Limit your contribution to about 2000 words
only.
Contact:
The Editor, The Indian Concrete Journal, ACC Limited,
L.B. Shastri Marg, Thane 400 604.
Tel: +91(22) 25825333, 25823631, ext. 653.
Fax: +91 (22) 25820962;
E-mail: editor@icjonline.com